---
dataset_info:
  features:
  - name: answer
    dtype: string
  - name: metadata
    dtype: string
  - name: prompt
    dtype: string
  - name: task
    dtype: string
  splits:
  - name: train
    num_bytes: 17135449.2
    num_examples: 9000
  - name: test
    num_bytes: 1903938.8
    num_examples: 1000
  - name: validation
    num_bytes: 856772.46
    num_examples: 450
  download_size: 7555741
  dataset_size: 19896160.46
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---